A reduced-reference perceptual quality metric for in-service image quality assessment
A reduced-reference perceptual quality metric for in-service image quality assessment
复制标题
DOI：
10.1109/tic.2003.1249092
复制
发表时间：
2003-12
期刊：
影响因子：
--
通讯作者：
T. M. Kusuma;Hans-Jürgen Zepernick
中科院分区：
文献类型：
--
作者：
T. M. Kusuma;Hans-Jürgen Zepernick
面向用户的图像质量评价已成为多媒体通信中的一个关键因素。然而，现有的图像质量指标，如峰值信噪比（PSNR）是不适合的服务质量监测，因为它们需要原始图像在接收器处可用。虽然PSNR和其他指标是客观的指标，但它们并不基于人类的视觉感知。为了克服这些问题，我们提出了一种新的减少参考客观的混合图像质量度量（HIQM），占人类的视觉感知，并不需要在接收机的参考图像。这种方法基于各种图像伪影测量的组合。结果是一个单一的数字，代表整体图像质量。实验结果表明，HIQM优于PSNR。
User-oriented image quality assessment has become a key factor in multimedia communications. However, existing image quality metrics such as peak signal-to-noise ratio (PSNR) are inappropriate for in-service quality monitoring, since they require the original image to be available at the receiver. Although PSNR and others are objective metrics, they are not based on human visual perception. In order to overcome these problems, we propose a novel reduced-reference objective hybrid image quality metric (HIQM) that accounts for human visual perception and does not require a reference image at the receiver. This approach is based on a combination of various image artifact measures. The result is a single number, which represents an overall image quality. Experimental results indicate that HIQM outperforms PSNR.